Lung cancer is the leading cause of cancer death in women. Although cigarette smoking has been established as the principal risk factor, an estimated 53% of female lung cancer cases worldwide are nonsmokers. The etiology of lung cancer in nonsmokers remains poorly understood. Postmenopausal hormone replacement therapy (HRT) has been shown to increase the risk of developing aggressive lung cancer and lung cancer mortality in two large clinical trials, whereas use of anti-estrogen tamoxifen is associated with a reduced risk of lung cancer death. These findings indicate that estrogens may play an important role in lung carcinogenesis. No study to date, however, has directly evaluated the association between endogenous estrogens and lung cancer risk. On the other hand, soy phytoestrogens, structurally similar to mammalian estrogens, possess anti- estrogenic activities. We recently found in a pilot study that soy food intake was significantly and inversely associated with both lung cancer risk and mortality. We also found that higher urinary levels of two soy isoflavones genistein and glycitein were associated with a lower risk of lung cancer in a small pilot study, and the inversion association was more pronounced for aggressive lung cancer. These findings are in direct contrast to the finding of a significant increase in lung cancer risk and mortality with HRT use in clinical trials. These findings, although preliminary, provide strong support for a novel hypothesis that soy isoflavones, possibly through their anti-estrogenic activities among other mechanisms, may favorably influence the development and progression of lung cancer. We propose in this application to evaluate the potential role of endogenous sex hormones and dietary phytoestrogens in the etiology of lung cancer in female nonsmokers. The proposed study will build upon the Shanghai Women's Health Study (SWHS), a prospective cohort of 74,941 Chinese women. Specifically, we propose to conduct a nested case-control study of 660 incident cases of lung cancer and their individually matched controls in nonsmoking postmenopausal women. Plasma sex hormones and urinary isoflavonoids will be measured in samples collected at baseline using liquid chromatography mass spectrometry based methods. In addition, we will examine the association of soy phytoestrogen intake with lung cancer risk and mortality in the full cohort. The SWHS represents an exceptional opportunity to address such targeted questions given its large size, prospective design, high follow-up rate, and availability of both urine and blood samples. More importantly, it provides repeated dietary assessments and a unique study population with an extremely low prevalence of tobacco product and HRT use but a wide range of soy food intake. Results from the proposed study will fill important gaps in our knowledge regarding the role of endogenous estrogen and dietary phytoestrogen exposure in the etiology of lung cancer and potentially contribute to the development of new strategies against this fatal disease.